Real-time vehicle overload detection method based on convolutional neural network

Abstract

The present disclosure provides a real-time vehicle overload detection method based on a convolutional neural network (CNN). The present disclosure detects a road driving vehicle in real time with a CNN method and a you only look once (YOLO)-V3 detection algorithm, detects the number of wheels to obtain the number of axles, detects a relative wheelbase, compares the number of axles and the relative wheelbase with a national vehicle load standard to obtain a maximum load of the vehicle, and compares the maximum load with an actual load measured by a piezoelectric sensor under the vehicle, thereby implementing real-time vehicle overload detection. The present disclosure has desirable real-time detection, can implement no-parking vehicle overload detection on the road, and avoids potential traffic congestions and road traffic accidents.

Description

CROSS REFERENCE TO RELATED APPLICATION[0001]This patent application claims the benefit and priority of Chinese Patent Application No. 202011526778.8, filed on Dec. 22, 2020, the disclosure of which is incorporated by reference herein in its entirety as part of the present application.[0002]TECHNICAL FIELD[0003]The present disclosure relates to an object detection technology, and in particular, to a no-parking truck overload detection method based on a convolutional neural network (CNN).[0004]BACKGROUND ART[0005]In road transportation, the overloading of trucks not only affects the safety of roads and bridges, but also poses great threat to people's lives in public. With loads higher than state-specified loads supported by the roads and bridges, the overloaded vehicles accelerate the loss of roads and bridges to cause massive maintenance funds, and often lead to traffic accidents. As the overloaded vehicles carry loads far greater than design loads of the roads and bridges, the roads...

AI Technical Summary

Benefits of technology

The patent describes a method for detecting the number of axles and wheels on a truck in real-time using a YOLO-V3 object detection algorithm and a pruning algorithm. The key to selecting the pruning channel is to search for a channel that contributes less to the output. By pruning channels based on their parameters and the proportion or threshold, the network model is simplified without affecting its detection precision. The method is desirable for real-time detection and can be used to avoid traffic congutions and accidents. Additionally, the method has a low hardware requirement, reducing device cost and being more suitable for various application scenarios.

Problems solved by technology

In road transportation, the overloading of trucks not only affects the safety of roads and bridges, but also poses great threat to people's lives in public.

With loads higher than state-specified loads supported by the roads and bridges, the overloaded vehicles accelerate the loss of roads and bridges to cause massive maintenance funds, and often lead to traffic accidents.

As the overloaded vehicles carry loads far greater than design loads of the roads and bridges, the roads where the vehicles are driven frequently are prone to pavement damage and bridge rupture, thereby greatly shortening the normal service life of the roads.

Such a manual detection method has the following defects: (1) The officers cannot work for 24 h; (2) on the basis of subjective judgments of the law enforcement officers, the precision cannot be ensured; and (3) the efficiency is low, and the vehicle-parking detection is time-consuming to cause traffic congestions easily.

Images